Skip to Main content Skip to Navigation
Theses

Traitement d'images et fusion de données pour la détection d'objets enfouis en acoustique sous-marine

Abstract : Detection and classification of buried objects is a particularly difficult problem: synthetic aperture sonar techniques used for seabed imagery provide data with very low signal to noise ratio, and then a large number of false alarms. The goal of this thesis is to make and develop algorithms allowing to reduce the number of false alarms, by keeping a good detection, and eventually to classify the detected objects, thanks to image processing and data fusion tools.For that, 1st, 2nd, 3rd, and 4th order statistical properties of these images are used in order to develop efficient detection algorithms. To improve the results, the data extracted by these means are combined by a fusion process using the theory of evidence. This allows to classify each pixel of the image into "object" or "not object" depending on wether it is supposed to belong to a sought object (underwater mine for example) or not. The result can then be used by an expert in order to help him in his decision.
Complete list of metadata

https://tel.archives-ouvertes.fr/tel-00011447
Contributor : Frédéric Maussang <>
Submitted on : Monday, January 23, 2006 - 4:57:55 PM
Last modification on : Friday, November 6, 2020 - 3:48:50 AM
Long-term archiving on: : Saturday, April 3, 2010 - 9:41:10 PM

Identifiers

  • HAL Id : tel-00011447, version 1

Collections

Citation

Frederic Maussang. Traitement d'images et fusion de données pour la détection d'objets enfouis en acoustique sous-marine. Traitement du signal et de l'image [eess.SP]. Université Joseph-Fourier - Grenoble I, 2005. Français. ⟨tel-00011447⟩

Share

Metrics

Record views

1311

Files downloads

1898